- CLONING
- SPEECH
- CONTENT
- HUMAN
Short answer
How can I make my AI sound like me?
You provide a clean audio sample of your voice to the Play.ht MCP. Your agent then uses that sample to generate a digital twin that can read any text you provide. You'll get a custom voice model ready for production.
The results of cloning
What you get from cloning
The process moves from a raw recording to a functional voice model.
CLONING
A digital voice twin
The AI processes your audio sample to map your unique vocal characteristics. You end up with a model that mimics your tone and cadence.
SPEECH
Instant text-to-speech
Once the clone is ready, your agent can turn any written text into audio. The output uses your new digital voice.
CONTENT
Voice-ready scripts
Your agent can draft text specifically formatted for your new voice. This ensures the speech sounds natural when generated.
HUMAN
Manual audio review
You still need to listen to the generated clips. You decide if the inflection matches your intent before using the audio.
The workflow
What your AI does when the audio arrives.
Your agent manages the heavy lifting of voice modeling and audio synthesis.
Analyze sample
The agent takes your uploaded audio file and extracts the necessary vocal patterns.
create_instant_voice_cloneBuild the twin
The AI creates a permanent digital representation of your voice in the Play.ht library.
create_instant_voice_cloneGenerate speech
Your agent sends text to the voice model to produce high-quality audio files.
create_instant_voice_cloneManage library
The AI organizes your various voice clones so you can switch between them easily.
create_instant_voice_clone
Try it
Copy these to start.
Use these lines to trigger the cloning process in your chat.
Starting points
These are prompts to get the process moving. Replace the bracketed text with your specific files or instructions.
Play.ht Connector
You're all set. Choose your MCP client and follow the setup instructions.
https://edge.vinkius.com/vk_preview_GOWw2EkNfCcipiaxRhjhVT8OcQmHdgozyQknlJBZ/mcpClaude Desktop
Follow the steps below to connect in seconds.
- 1In Claude Desktop, open Settings → Connectors.
- 2Click “Add custom connector” and paste the connector link above as the remote MCP server URL.
- 3Click Add and start a new chat — Play.ht capabilities are ready to use.
{
"mcpServers": {
"playht-voice-cloning-mcp": {
"url": "https://edge.vinkius.com/vk_preview_GOWw2EkNfCcipiaxRhjhVT8OcQmHdgozyQknlJBZ/mcp"
}
}
}Use my uploaded audio file to create an instant voice clone.
Create a voice clone using the recording I just shared in this chat.
Now that you've cloned my voice, read this script using the new model.
List all the voice clones you have created for me so far.
Claude
ChatGPT
Cursor
VS Code
Windsurf
Claude Code
JetBrains
Cline
Start here
Connect Play.ht once, then ask.
Click the link to connect your account. Your credentials stay encrypted, and you can start cloning voices immediately through your preferred AI client.
Connect Play.ht to your AIFAQ
How this task behaves
- 01
How long does it take to create a clone?
The AI processes the sample and builds the model immediately after you provide the audio.
- 02
Can the AI clone any voice from a video?
No. You must provide a clear audio sample for the tool to work effectively.
- 03
Does the AI decide which voice to use?
No. You must tell your agent which specific voice clone to use for a text-to-speech task.
- 04
Can the AI edit the audio files it creates?
No. The tool generates the speech; you must use separate audio editing software if you want to change the output.
- 05
Will the AI delete my original audio files?
No. The tool only uses the audio to build the model; it does not delete your source files.
More questions about Play.ht? The Connector page answers them. See everything the Play.ht Connector can do
Connect Play.ht to Claude, Cursor, ChatGPT & more
